Abstract
We develop an interface system using hand gestures to operate electrical appliances intuitively in a daily life scene. Hand-waving detected in the image is used as a cue to start the system, to distinguish gestures based on the user's intension of operation from other daily-life motions, and to specify the position of the user's hand. Hand-waving detection is implemented based on combination of dynamic background subtraction method and skin color extraction. In addition, we focus on changes of brightness value in each pixel. The user selects a target appliance to operate by pointing out with a hand under a ceiling camera with fisheye lens. The appliance within the region that the user indicates is decided as target one. A PTZ camera is also used to zoom in the user's hand area and to observe hand gestures. The shape of the hand is basically recognized using HLAC features. We add a size of hand region as a new feature to the HLAC and employ a coarse-to-fine strategy using image pyramid. Robustness and the success rate of gesture recognition, consequently, are improved. In this paper, we prepare seven types of hand shapes to operate the channel of TV. Effectiveness of the proposed system is shown through experiments.
